Code is usually over specified. I recently used AI to build an app for some HS kids. It built what I spec’wd and it was great. Is it what I would’ve coded? Definitely not. In code I have to make a bunch of decisions that I don’t care about. And some of the decisions will seem important to some, but not to others. For example, it built a web page whereas I would’ve built a native app. I didn’t care either way and it doesn’t matter either way. But those sorts of things matter when coding and often don’t matter at all for the goal of the implementation.
It may be unreliable to you. I see the life of most people around me getting better. Even people that are somewhat poor (not dirt poor, but free lunch poor) have homes, three squares and snacks, PS5, mobile phones with cellular data, and cable tv. The biggest life issues I see are usually strongly related to substance abuse and mental health.
Transplanting to even just the 80s would be a culture shock for most people.
Really curious. How is life getting better for most people? No one has kids, they can't afford families, medical care is unreachable, housing is unaffordable, I guess if your life is scrolling on an iphone its MUCH better but if you want to live, educate a family, retire, live in a safe community, or have healthcare its worse in every measure.
There is huge variation in what the US trend looks like from the ground that varies by region, age, income level, industry, and demographic.
EI think if you’re a professional class baby boomer the trajectory has looked fantastic through your life.
If you’re a 35 middle income living on the coasts (where at least 100 million Americans live) you may have watched affordability collapse and QOL decease significantly over the last decade.
Allergies are not simply overactive immune response. It’s the wrong type of response. What’s really intriguing is how much we can do innate immunity that we have done relatively little with.
> His first mistake was complaining to HR about another employee griefing him. HR is always going to consider the initial complainer as "the problem."
I can say this definitely isn't always true. In the companies I've worked at HR has always been extremely reasonable and cooperative with harassment claims. But corporate culture probably matters here, and I've never worked at a place like Uber.
That said, I would be curious to actual know the correspondence that was sent between the two. I can also say being a manager who has had to deal with a situation between two employees (more than once), they often both claim to be the one being harassed -- and usually even a little bit of digging reveals really clearly who the aggressor is.
The phrasing "HR isn't there to protect you, it's there to protect the company" applies more here.
My experience is also that HR is very reasonable and cooperative with harassment claims. But the thing is that when you have a legit harassment claim, the law is there to protect you. You could make things very expensive for the company in court, and so protecting the company does mean protecting you and treating you respectfully and cooperatively.
If HR investigates and finds you don't have a legit case and that in fact you may have been the instigator, then protecting the company probably means getting rid of you. Your judgment and account of the facts is questionable in that case, and you're a liability from the other side.
I don't know exactly what happened in this case, but in the harassment case I've had to handle as a manager, the (male) employee said that the (female) victim had initiated everything and had this weird fascination with him, while the paper trail that everybody could see clearly showed that he was both the instigator and the one behaving improperly. Projection is strong in cases like these. So it's entirely possible we're not getting the full story from this anonymous blog post.
> and that in fact you may have been the instigator, then protecting the company probably means getting rid of you.
That protects other employees. If you are instigator and then go to complain to HR trying to make them punish the victim, firing you protects everyone around you. And it protects the culture from becoming toxic.
HR can play negative role, but this scenario is not one of those.
> The phrasing "HR isn't there to protect you, it's there to protect the company" applies more here.
I agree (although had interpreted the statement originally differently). Unfortunately, the part about "XYZ isn't there to protect you" applies to so much in life. Even police don't have a responsibility to you protect you (but just the public as a whole). The lesson from stuff like this is often to make sure your best interest are aligned with the most powerful and active stakeholder in the "room".
Or don't engage with people whose interests are not aligned with yours. You can do an awful lot, and carve out a pretty good life for yourself, if the powerful people whose interests are not aligned with yours don't know that you exist. Considering that everybody else has an incentive to align with the most powerful and active stakeholder in the world, this is the only way to avoid a unipolar dictatorship.
Relating it back to the story at hand, the blogpost's author would've done well to just disengage from the coworker who didn't like him, and also to not report them to HR. What I had to tell my report when HR got involved: "The right thing to do here was nothing."
Wait -- I'm fairly certain this is obvious to the person you were responding to. It may not be obvious to a lay person (who may not even know LLMs are trained at all). But I think this is obvious to almost all people with even a small understanding of LLMs.
I'm actually pretty convinced they're a troll or at the very least a high confrontation participant who is quick to move goal posts, ignore entire chains of logic, engage in ad hominim attacks of other posters, and is bringing zero novel insight anywhere in this thread
one of the posters said it can't even reason through chess, i ran the actual benchmark, spent money and actually proved that it can beat a 1000 elo chess engine.
The most frustrating part for me is that this is how I used to write. I was always doing, "Why X works, but Y doesn't" and stuff like that. I may have seemed trite or pompous (or both) in the past, but now it seems like I'm copying an LLM -- which actually feels worse. One thing I haven't seen ChatGPT do much of is use sound-effects, so swoosh here we go with my new writing style schwing!
I feel you. I've been using en-dash in my writing for decades, but finding myself removing them now for fear of being mistaken for an LLM. (They tend to use em-dash, but I don't think people are going to distinguish between – and —.)
Do you think pre-AI writing is going to become really valuable because it is free of any AI assistance? If we all start using AI to assist in writing, then pre-AI writing may become important, similar to pre-atomic steel (i.e., https://en.wikipedia.org/wiki/Low-background_steel)
>Do you think pre-AI writing is going to become really valuable because it is free of any AI assistance?
Serious question: Do you think old pictures are valuable because they are free of photoshop? Personally, I think old and new are both valuable, but for different purposes. Technology gave us new capabilities with new hope
Not the user you asked, but: Yes, it seems obvious that old pictures are valuable because they are free of photoshop. Not that this means that they are free of manipulation though, c.f. the famous picture of stalin at the river with/out his fellows.
I predict in the future a humans.txt for each site that indicates the level of human authorship and for fully human authored content to be highly valuable
It's already important, for various reasons. E.g. I love older electronics books where they explain things in a very thorough manner (maybe because they had more time?). But of course reading older books is full of traps, in some subjects you need to be more careful than when analyzing the output of an LLM.
That is what I mourn the most. They were my punctuation get-out-of-jail free card.
I didn’t love them enough to figure out how to type them without doing two dash’s in Word and then backspacing out of one and hitting space again — but damnit, I miss it.
Before the LLM craze I didn't even know — was specifically different than just -, and I used it in the same way. But now I notice specifically when people use either, and when people use -- instead.
I would think to most people, (myself included!), it's just a 'dash'. A sentence was written with a dash - you could just ingest and read past it, like a comma.
Not saying this is accurate usage, maybe just real world usage.
I would hope most people can distinguish between the really short dash and the longer forms, even if they don't know any of the rules around them. But n versus m I don't expect people to notice.
I’m not sure I’m representative of “most people” in this respect (I have always used both n and m dashes), but I personally find the difference between n and m dashes bigger and more noticeable than the difference between regular and n dashes.
Because most people are ESL and really don't care.
I didn't even know there are multiple types of dashes.
I did know about multiple types of quotes because they kept breaking code on blogs. Still didn't care, but at least I learned how to spot and fix them.
Really looking forward to having the wrong kind of dash in code, but at least with current tech that seems like it won't happen.
Why wouldn't they. Never studied them. Never even thought twice about the dashes in a sentence. Didn't realize they were different till like a few months ago when everybody suddenly started focusing on how "AI" it makes everything look
And of course, the reason that ChatGPT sounds like that is that it's what a whole lot of explanatory expert blog posts did, and so when ChatGPT is told to talk like that, that's what it does.
So interesting! There was a level of creativity back in those days that we seem to not have as much nowaydays. Now things seem to be based more on math and things like signing.
As a C64 user a kid one thing I didn't realize was this:
"The serial bus connecting the 1541 to the C64 has a famous bug. Commodore designed it to run at ~16,000 bytes/second, but a timing error in the Kernal ROM's ATN (Attention) line handling caused it to insert extra delays between every single byte — dropping the real-world speed to around 400 bytes/second. That's roughly one-fortieth the intended speed. Commodore never fixed it in the C64's lifetime. (A workaround was eventually added for the 1571 drive, but the 1541 remained slow throughout its production run.)
Fast loaders worked around this by downloading replacement code into both the drive's RAM and the C64's RAM, then communicating over the C64's User Port parallel pins instead — bypassing the serial bus completely. This gave transfer rates around 10,000 bytes/second (25× faster). "
I recall the "fast loaders" and always wondered how they could get data off a disk any faster. It seemed like magic. But now I know.
It's worse than that even: there are delays after every _bit_ transferred, due to various design decisions that were made along the way, which is why the C64's disk drive is slower than the previous computer (the VIC-20), which is slower than the one before that (the PET).
I wrote about the decisions and the resulting delays for one of those 100-post Threadapalooza projects in 2024, compiled here for easier reading: https://imrannazar.com/articles/commodore-1541
Not buying that, C64 had like 5-10 pcb revisions so spinning another one wouldnt be extraordinary, and in the mean time they could put bodges on old stock pcbs or you know, supply USERPORT cable as thats where 6526 is wired to. Original Kernal has no traces commodore ever tried to use hardware shift register, they simply left VIC20 bodge and didnt even try accommodating fixed 6526.
> There was a level of creativity back in those days that we seem to not have as much nowaydays. Now things seem to be based more on math and things like signing.
Copy protections nowadays are actually extremely complex - just look at Denuvo and VMProtect. I presume that nowadays there are less copy protection schemes because producing a resilient one is too complex for small developer teams.
That part about fast loaders bypassing the serial bus and using the user port is incomplete at best, incorrect in practice. Some early fast loaders did this but most others just used the existing serial bus connection with their own routines, bypassing the buggy kernal drivers on the drive and host side.
There are more errors in the text, e.g "the 1541 used a single-density (narrow) read/write head". No, that describes an 80-track head while the 1541 used a wider 40-track head. I stopped reading here because of these errors.
OK, someone needs to write up an accurate version. This is a fascinating topic, especially for those of us where C64 was one of our first computers. The Vic20 was my first-- I sold my computer games door-to-door on cassette... "sold" is maybe a strong word here. I sold one total.
The tape copy protection article is also terrible with nonsense like this "A copy of a Novaload tape made on a HiFi deck produced a normal-speed recording with the correct audio signal but which, when loaded on a C64, produced garbage — because the Kernal's standard tape decoder couldn't interpret the turbo encoding"
Thousands of kids in the playground with access to a tape-to-tape deck (which made Alan Sugar his fortune for Amstrad) would disagree. And the reason given is just plain wrong.
A plan? Actually there is. This is all part of the backdrop to end US elections. We can’t have elections in the middle of a major war. And if we do have them we must greatly constrain how they are held while we are at war.
We had elections during WW2, the largest war of all time; we had elections during the civil war when confederate troops were 30 miles from DC. An air campaign in the Middle East is just another tuesday by comparison. This theory falls flat on its face - it is not a reasonable pretext for suspending elections, and this administration does not bother with creating pretexts for its power grabs.
Ah, but whether it is a "reasonable" pretext/excuse for suspending elections is up to the media and how they want to spin it for the masses, to shape their opinion, isn't it? And how practical, that more news outlets are now owned by MAGA people. Furthermore, I will not put it past Trump to use any flimsy excuse to suspend elections, if he thinks he will lose.
What if Trump were to say elections are illegal due to the war. We need to delay them. And Republicans in congress did nothing. And the Supreme Court decided not to hear any cases related to it. What then? We’re learning the US government has basically no teeth to stop something like this.
It's a scary thought, albeit not a realistic one at the moment, thankfully. The Supreme Court has shown ample willingness to strike down blatant (and subtle, for that matter) executive overreach. Exhibit A is Trump's tariffs, which were justified by the administration to be legal through the International Emergency Economic Powers Act, which allows the president to “regulate…importation” during a declared state of emergency. The Supreme Court found that the wording in the act allowing the president to “regulate…importation” was not sufficient to grant the president the power to impose tariffs. The wording in the IEEPA is vague enough that you could go either way, but the conservative majority tends to follow the Major Questions Doctrine, which essentially says that in vague matters like this, assume that the power belongs to Congress and not the president.
Meanwhile, delaying or canceling elections through executive order would be blatantly illegal, particularly when no conflict is taking place on U.S. soil. The case likely wouldn't even make it to the Supreme Court, but if it did, I have no doubt elections would be promptly reinstated.
I'm not saying the Supreme Court has a perfect record, of course. Not even two years ago, they essentially ruled that the president is above the law. But at least in matters regarding the balance of powers between branches, the Supreme Court is wary of the power of the executive branch, and that should certainly include the president's ability (or lack thereof) to interfere in elections.
Can you name something which can't be spun by the media or that you could not believe Trump would try to use as an excuse? If something is always true, it is evidence of nothing in particular.
Claiming this strike on Iran is an attempt to suspend US elections is exactly as ridiculous as claiming the last round of strikes on Iran, or the Maduro raid, or any of Trump's other previous military boondoggles were attempts to suspend US elections.
I've been trying to avoid the news for a little over a year now. I needed a detox. ... Is this true? That is, are there legitimate proposals to cancel or constrain the November elections in any significant way? Or, is this all speculation?
There's a memo out about nationalizing elections and there's the SAVE America act to require much stricter voting requirements. Both of these unconstitutional obviously because federal government doesn't run elections.
Isn't this essentially what MAGA argued during the fighting over the 2020 election - that the states should be able to run their elections however-the-fvck they want, and the feds have no right tell the states how to run their elections?
When zelenskyy mentioned elections were suspended by the war to trump, in the Whitehouse while in a room full of media, trump replied something like "now that's a good idea"
> We can’t have elections in the middle of a major war.
Yes we can? Is there any provision in the US Constitution that allows delay of election because of war? We have had elections during most of our recent wars (Iraq, Vietnam, Korea, Afghanistan).
Trump could definitely try. Or pull an emergency card out of his ass. But it doesn't mean there is any provision for cancelling elections because of this 'war' with Iran (which they aren't even calling a war, but a "special combat operation" to get around congress having the war powers)
In 1995 no one used the web still. Sure, we all did, but it was pretty niche. I think you could argue that chatbots are niche as well, but the user base of OpenAI is way larger now than Netscape in 1995. Netscape had probably 25 million users at the end of 1995. ChatGPT has about 800 million.
reply